Visual deficits and retinotoxicity caused by the naturally occurring anthelmintics, Embelia ribes and Hagenia abyssinica

TOXICOLOGY

AND APPLIED

PHARMACOLOGY

81,220-230

(1985)

Visual Deficits and Retinotoxicity Caused by the Naturally Occurring Anthelmintics, Embelia &es and Hagenia abyssinica’ G. Low,*

L. J. ROGERS,* S. P. BRUMLEY,# AND D. EHRLICH~

Departments of fAnatomy and *Pharmacology, Monash University, Clayton, Victoria 3168. and $AIfred Hospital, Commercial Road, Prahran, Victoria 3181, Australia

Received January 16, 1985; accepted June 4, I985 Visual Deficits and Retinotoxicity Caused by the Naturally Occurring Anthelmintics, Embelia ribes and Hagenia abyssinica. LOW, G., ROGERS, L. J., BRUMLEY, S. P., AND EHRLICH D. ( 1985). Toxicol. Appl. Pharmacol. 81, 220-230. The naturally occurring anthelmintics, Embelia ribes (Enkoko) and Hagenia abyssinica (Kosso), have been reported to possibly cause optic atrophy among the Ethiopian population. In this study we found retinal pathology and defects in visual behavior in chicks treated with Enkoko, Kosso, or embelin, a crystalline extract of E. ribes. The chicks were fed orally with a high dose of 0.25 g (5 g/kg) or a low dose of 0.025 g (0.5 g/kg) per day of Enkoko or Kosso, beginning on Day 2 of life. The high dose for Enkoko was administered for 1 or 5 days, while that for Kosso was administered for 1 or 9 days. For the low dose of both Enkoko and Kosso, the dosing regime was for a period of 1,4, or 9 days. Embelin was administered at a dose of 0.001 g (0.02 g/kg) per day for 9 days. Control chicks were force fed an equivalent amount of chick feed. Treatment with Enkoko or Kosso significantly reduced the ability of chicks to detect a moving bead introduced into the peripheral field of vision. The degree of constriction of the visual field for detection was dependent upon the total amount of drug administered. Performance on a visual discrimination task, which required discrimination of feed grains from pebbles, was also impaired in chicks treated with total doses of 0.200 and 0.250 g of Enkoko or Kosso. Thus, the extent of deficit in visually guided tasks was found to be dose dependent. The visual deficits observed in Enkoko-treated chicks were mimicked by embelin, which suggests that embelin may be responsible for the visual defects. Anatomical evidence of degeneration of ganglion cells was found in retinae exposed to high doses of Enkoko ( 1.25 g) and Kosso (2.25 g). However, no retinal lesions were detected in chicks following treatment with cumulative doses of less than 0.25 g of Enkoko or Kosso. Similarly, retinal pathology was not observed in embelin-exposed retinae. 0 1985Academic yes, IOC.

The naturally occurring anthelmintics, Embelia ribes (Myrsinaceae) and Hagenia abyssinica (Rosaceae) are used commonly in East Africa where they are known locally as Enkoko, or Umkoko, and Kosso, or Kousso (Gupta et al., 1977; Pankhurst, 1979), respectively. Kosso is the female flower of H. abyssinica , and Enkoko is the seed of E. ribes. As well as being used to treat helminthic infec’ Part of this work has been communicated at the Proceedings of the Australian Physiological and Pharmacclogical Society, 1984. 0041-008X/85

$3.00

Copyright Q 1985 by Academic Press, Inc. All rights of reproduction in any form mrved.

tions, both Enkoko and Kosso have been used as abortifacients in traditional medicine (Casey, 1960; Farnsworth et al., 1975a,b; Seshadri et al.. 1978). This has led to a number of studies investigating Kosso and Enkoko, plus one of the latter’s crystalline extracts, embelin, as possible contraceptive agents (e.g. Chaudhury, 1966; Radhakrishnan and Alam, 1975; Purandare et al., 1979; Dixit and Bhargava, 1983). Since Enkoko and Kosso are still widely used in Africa today and are being investigated as possible contraceptive agents in the future, 220

RETINOTOXICITY

CAUSED BY ANTHELMINTICS

it is of some importance to determine whether they have any toxic effects. Rokos ( 1969) and Brumley (unpublished results) have suggested that ingestion of Enkoko or Kosso may cause optic atrophy in susceptible subjects. In the limited study by Rokos (1969), retinopathy, presumably caused by these anthelmintics, consisted of a progressive restriction of visual field resulting in complete blindness by 4 to 6 weeks after a single dose (8 to 16 g) of the dried flowers/seeds of either of the drugs. He attributed this effect to overdosage of the drugs or to individual hypersensitivity. In this study, we investigated the possible toxic effects of ingested Enkoko, Kosso, and embelin on visually guided behavior in chicks, followed by histological examination of their retinas. Chicks are suitable subjects for such anatomical studies since they have relatively large eyeballs containing well-defined retinal layers. All cell types found in the mammalian neural retina are present in the chick retina (Rodieck, 1973). Furthermore, the chick retina does not contain any blood vessels traversing the vitreal surface (Ehrlich, 198 1) which may complicate the interpretation of any pathological changes. In addition, chicks are excellent subjects for testing on visually guided behavioral tasks. METHODS Animals and housing conditions. Male Australop-Leghorn cross chicks (40 to 45 g) were obtained from a local hatchery on their first day of life. They were housed in groups of five until Day 4 posthatch. Thereafter, they were visually isolated from each other. Housing consisted of rootless, metal cages(23 X 23 X 30 cm) with sliding Perspex fronts. Constant light and warmth were provided by a 25W globe suspended above the cage. Grains of chick starter crumbs and water were available ad fibitum. Drugs and drug treatment. Enkoko and Kosso were administered orally in crude form commencing on Day 2 of life. The dried Enkoko seedsor Kosso flowers were ground and made into a paste with distilled water (ratio of volume of drug to water, 1:3). Crystalline embelin was mixed with chick starter crumbs and also made into a paste (ratio of volume ofembelin to starter crumbs to water, 1:2:4). Controls were fed with a paste of starter crumbs (ratio of volume of starter crumbs to water, I :3). Two different sized doses of the drugs were used; a high dose of 0.25 g (5 g/kg) per day and a low dose of 0.025 g

221

(0.5 g/kg) per day. The high dose was administered for either 1 or 9 days in the case of Kosso, and for either 1 or 5 days in the caseof Enkoko. The dosing period for Enkoko was shorter since chicks became ill and were therefore killed on Day 6. The low dose was administered for 1,4, or 9 days for both Enkoko and Kosso. Crystalline embelin was given at a dose of 0.001 g (0.2 g/kg) per day for a duration of 9 days. There was a control group corresponding to each treatment condition. Each group consisted of an average of eight subjects. Behavioral testing. Animals were tested for behavioral deficits in visual function and subsequently used in the anatomical investigations. With the exception ofthe highdose Enkoko group, which appeared to be suffering from nonspecific toxic effects, all testing was performed according to a randomized blind procedure. Two visual tasks were used. Visual field detection: Perimetry, which involves the mapping of visual field, is a standard and useful procedure for diagnosing visual pathology in humans (Merigan, 1980; Potts, 1980). The field of vision is measured by moving a target of standard size and brightness from the periphery toward the center until it is detected by the subject. The visual field detection task used in this study follows the same principles. It requires detection of a stimulus introduced into a chick’s peripheral field of vision. The stimulus is a red bead (2 mm diameter) attached to the end of a thin wire. The bead is manually advanced toward the chick’s beak in an arc of radius 5 cm, along a horizontal plane at the approximate level of the chick’s eye. When the chick detects the bead. it gives a startle response which is sometimes accompanied by a trill vocalization. This overt reaction to the bead is characterized by the chick giving a quick startle jerk, followed by rapid turning of the head toward the bead. The chick almost always pecks at the stimulus. The position of the bead on its detection is scored as the angle from the long axis of the chick’s beak (Fig. 1). This angle was measured to the nearest 10”. Each chick was tested daily; its angle of detection was determined three times consecutively for each eye, and the average score was taken, In order to ascertain inter-rater reliability, two experiments independently made their judgements in three separate experimenters which involved the testing of 54 chicks. Both their estimations correlated well (r = 0.88), indicating that there was 88% agreement between the two experimenters. Visual discrimination learning: This test requires visual search for food grains scattered on a background of pebbles adhered to a Perspex floor (Rogers et al., 1974). It is a task which predominantly uses the frontal or binocular field. Each chick was tested on Day 9 of life, after 3 to 4 hr deprivation of feed. A total of 60 pecks was allowed in the task. The number of errors (i.e., pecks at pebbles) scored in the last 20 pecks was taken as a measure of the ability to perform the visual discrimination task. Microscopy. Chicks were killed with an overdose of Nembutal(60 mg/ml sodium pentobarbitone, ip) on Day

222

LOW ET AL.

nificant decrease in ability to detect a small bead moved into the peripheral field of vision [F&24) = 6.07;~ < O.OOl] (Fig. 2). Beginning on Day 3, there was a significant reduction in the size of the angle of detection in the Enkoko [F(4,30) = 30.73, p < O.OOl]- and Kosso [F(4,30) = 17.06, p < O.OOl]-treated chicks. This visual angle of detection decreased with each day of treatment of Enkoko or Kosso, across Days 3, 4, 5, and 6 posthatch [F(2,30) = 11.82, 23.13, 3 1.29, and 54.20, respectively; p < 0.00 I]. The chicks treated with 0.25 g Enkoko per day were very ill by Day 6 of life and would not have survived much longer. Thus MOVEMENT OF THE STIMULUS a total amount of 1.25 g of Enkoko appeared EITHER INTO THE RIGHT OR LEFT PERIPHERAL FIELD OF VISION to be lethal to chicks and, by Day 6, this treatment had significantly reduced the angle of FIG. 1. A schematic illustration of the angle of detection measured in the visual field detection task. The axis of the detection to about 20”. An equivalent amount beak defines 0”; all angles to the left are defined as the left of Kosso was not lethal. It is interesting to note visual field (LVF) while those to the right are defined as that although drug administration continued the right visual field (RVF). For any given head position, until Day 11, the visual angle of detection of the angle between the axis of the beak and the stimulus at a central point in between the two eyescan be estimated. chicks dosed with 0.25 g of Kosso for 9 days This angle, (Y, is the angle of detection. remained at a mean angle of 35” from Day 6 onward [F(6,24) = 1.02, p > 0.051. When Enkoko or Kosso was administered 14 of life, except for those receiving the higher dose of Enkoko (0.25 g X 5 days), which were killed on Day 6 of at a dose lo-fold less than 0.25 g for a similar life. The eyes were quickly excised and immersed in a duration (i.e., 0.025 g X 9 days), there was a mixture of 5% glutamldehyde and 1.2% pamformaldehyde in 0.1 M sodium cacodylate buffer (pH 7.4). Samples of significant interaction of treatment effects retinal tissue (2 X 2 mm) from the central, nasal, and acrossdays [F(14,105) = 3.91,~ < O.OOl] (Fig. temporal regions were postfixed in 1% osmium tetroxide, 2). Further analysis of this interaction revealed dehydrated in a graded series of ethanol, and subsequently that the mean visual angle of treated chicks embedded. Transverse semithin sections of 1 pm were significantly decreased over Days 8 [F(2,120) stained with toluidine blue and examined with a light mi= 6.24, p < 0.011, 9 [F(2,120) = 9.69, p croscope. < O.OOl], and 10 [F(2,120) = 14.47, p Statistical procedures. Parametric statistics (Keppel, 1982; Winer, 197 1) were used to analyze the behavioral < 0.00 I]. A two-factor (treatment X days) data. In the visual field detection task, a three-way analysis ANOVA revealed that the mean visual angle of variance (ANOVA) for repeated measures was used to of detection of Enkoko [F(2,120) = 1.50, p determine the effects of drug administration on the daily > 0.05]- and Kosso (F < 1.OO)-treated chicks angle of detection of the chicks. Further analysis with simple effects was carried out where appropriate, in order to plateaued from Day 12, and the final angle of locate the exact source of interaction. In the visual disEnkoko-treated chicks was about 20” whereas crimination task, independent t tests were used to assess the difference in errors in the last 20 pecks made by each this angle was 60” for those treated with Kosso. Thus, a dose dependency is evident in both treatment group. the Enkoko- and Kosso-treated chicks on this RESULTS task. Figure 3 presents data from another experVisual Field Detection iment where both dose and duration of treatTreatment with Enkoko (0.25 g X 5 days) ment of Enkoko or Kosso were varied. There and Kosso (0.25 g X 9 days) resulted in a sig- were significant effects of treatment across days

RETINOTOXICITY

CAUSED

223

BY ANTHELMINTICS

c 14 13 12 11 10 9 6 7 6 5 4 3 2 120

100

LEFT

60

VISUAL

60

40

Z"

20

FIELD DETECTION

40

60

RIGHT ANGLE

FROM

VISUAL

60

100

120

FIELD

BEAK

FIG. 2. The constriction of visual fields in chicks after oral administration

of Enkoko and Kosso. The horizontal axis represents the angle of detection (degrees) when the stimulus is introduced either into the left or right visual field. The vertical axis represents the age of the chick in days (posthatch). Symbols represent mean visual angle of detection (&SE) of control (0), and Enkoko (A), and Kosso @)-treated chicks at doses of 0.25 g/day (-) or 0.025 g/day ( * * m).Drug administration commenced on Day 2 of life for a duration of 9 days. The only exception was the higher dose of Enkoko, where the chicks were killed after 5 days of drug administration.

for both Enkoko [F(3,16) = 7.29, p < 0.0 I] and Kosso [F(3,20) = 11.74, p < O.OOl]. Furthermore, the visual angle of detection decreased as a function of the total amount of Enkoko (i.e., 0.025,0.10, and 0.25 g) [F( 1,321 = 11.94, 8.29, and 34.29, respectively, p < 0.011, or Kosso (i.e. 0.025,O. 10, and 0.25 g) [F( 1,40) = 14.68, 62.02, and 106.56, respectively, p < O.OOl]. Embelin given at a dose of 0.001 g per day for 9 days also significantly reduced the angle of detection across days [F( 1,ll) = 15.96, p < 0.011 (Fig. 4). It is interesting to note that the pattern of constriction of visual angle is similar to that of chicks treated with a total dose of 0.25 g of Enkoko.

paired discriminatory ability (Fig. 5). This deficit was significant at higher doses of both Enkoko and Kosso. Compared to controls, chicks treated with a total amount of 0.200 g [t(78) = 2.39, p < 0.011 and 0.25 g [t(72) = 3.59, p < O.OOl] of Enkoko made significantly more errors than did the controls. Those treated with Kosso at 0.200 g [t(78) = 1.83, p < 0.051 and 0.25 g [t(73) = 3.48, p < O.OOl] also made significantly more errors than the controls. Likewise, embelin (0.001 g X 8 days) significantly impaired the chicks’ performances on this discrimination task [t(68) = 3.64, p < O.OOl].

Visual Discrimination

Dissection of eye cups exposed to Enkoko, Kosso, and embelin revealed a normal, undeformed, translucent lens. The ciliary body and vitreous also appeared normal. An increase in size of the eyeball is a sen-

Task

In the visual discrimination task where chicks had to discriminate grains from a background of pebbles, the treated chicks had im-

Retinal Examination

LOW ET AL.

224 D 14 13 12 11 10 9 6 7 6 5 4 3 2 120

100

60

LEFT

60

40

VISUAL

FIELD

20

20

40

RIGHT

DETECTION

ANGLE

FROM

60

VISUAL

60

100

120

FIELD

BEAK

D 14

(b1

13 12 11 10 9

I 120

100

60

LEFT

60

VISUAL

40

20

FIELD DETECTION

0

20

40

RIGHT ANGLE

FROM

60

VISUAL

60

100

120

FIELD

BEAK

FIG. 3. The effect of varyingdoseand duration of treatment of Enkoko (a) and Kosso(b) in the visual field detectiontask.Data are plotted asin Fig. 2. Symbolsrepresentmean visualangleof detection(*SE) of Enkoko-or Kosso-treatedchicksat three differentdose.regimens:0.025 g X 1 day (0); 0.025 g X 4 days (a); 0.25 g X I day (0). The correspondingcontrol groupswerecollapsedand are representedas (0).

sitive criterion for raised intraocular

pressure

in young birds (Wallman and Turkel, 1978; Yinon et al., 1980). This increase was not observed in any of the treated animals although direct measurements of intraocular pressure

were not taken. The light microscopic examination of retinas from control and treated chicks revealed morphological changes in the ganglion cell layer (GCL) of the Enkoko- and Kosso-exposed retinas. The eyes of 11 En-

RETINOTOXICITY

120

100

80

LEFT

VISUAL

60

40

CAUSED

20

20

FIELD DETECTION

40 RIGHT

ANGLE

225

BY ANTHELMINTICS

FROM

60 VISUAL

80

100

120

FIELD

BEAK

FIG. 4. The constriction of visual field in chicks after oral administration of 0.001 g/day (9 days) of embelin. Data are plotted as in Fig. 2. Symbols represent mean visual angle of detection (&SE) of control (0) and embehn (@)-treated chicks.

koko-treated (0.25 g X 5 days) and 9 Kossotreated (0.25 g X 9 days) chicks were examined. The initial effects of Enkoko and Kosso observed in the three retinal regions appeared to be a specific and severe destruction confined cn

to the ganglion cells. All regions sampled appeared equally affected. Ganglion cell nuclei were pyknotic and were associated with hyperchromatic hyperplastic cytoplasm (Figs. 6b and c). Retinas from chicks treated with low

c

CONTROL

ENKOKO

KOSSO

EHBELIN

5. Impairment of visual discriminatory ability in chicks treated with Enkoko, Kosso, or embelin. The chicks were required to discriminate feed grains from a background of pebbles. The number of errors in the last 20 pecks is plotted on the vertical axis. Data are represented as group mean @SE). The doses used were 0.025 g (M), 0.10 g @II),0.200 g(o), and 0.25 g (@ for Enkoko- and Kosso-treated chicks. Embelin (a) was administered at a total dose of 0.008 g. Significance levels for comparison to the appropriate control group: *p < 0.05; **p < 0.01; ***p < 0.001. FIG.

226

LOW ET AL.

a

FOG.6. Light micrographs of chick retina taken from equivalent retinal sites. (a) Control retina showing lamination pattern of nuclear and plexiform layers. OFL, optic fibre layer; GCL, ganglion cell layer; IPL. inner plexiform layer; INL, inner nuclear layer; OPL, outer plexiform layer; ONL, outer nuclear layer; P, photorecptor layer. Cells in the GCL project their axons in the OFL, which ultimately collects to form the optic nerve. (b) Enkoko-treated retina (0.25 g X 5 days). Note the pyknotic cells in CCL (arrowed) and the apparent reduction in their numbers. (c) Kosso-treated retina (0.25 g X 9 days). Note the similar pyknotic appearance of cells in CCL (arrowed) as in (b), but the apparently normal number of ganglion cells remaining in this layer. (d) Embelin-treated retina (0.001 g X 9 days). Note that cells in the CCL appear essentially normal in spite of the behavioral changes observed. Bar = 50 pm. Toluidine blue stain.

RETINOTOXICITY

CAUSED

doses (0.025 g X 1 day; 0.025 g X 4 days; 0.025 g X 9 days; 0.25 g X 1 day) of Enkoko or Kosso looked essentially normal. Pyknosis of ganglion cells was also not evident in embelinexposed retinas (Fig. 6d). Other layers of the retinas, that is, the optic fibre layer (OFL), inner plexiform layer (IPL), inner nuclear layer (INL), outer plexiform and outer nuclear layer (OPL and ONL), and photoreceptor layer (P), revealed no qualitative changes compared with normal retina.

BY

ANTHELMINTICS

227

(0.025 g X 8 days) and 0.25 g (0.25 g X 1 day) of either Enkoko or Kosso. The treated chicks were less able to discriminate feed grains from a background of pebbles than could controls. However, lower doses, which significantly impaired performance on the moving bead detection task, failed to affect visual discrimination. It is important to stress that, whereas the bead detection task detects deficits in the peripheral field of vision, the visual discrimination task primarily uses the binocular field vision. It is therefore important to note that, while chicks treated with high doses of Enkoko DISCUSSION or Kosso could detect the bead when it entered The results indicate that the naturally oc- the binocular field of vision, their vision was curring anthelmintics, E. ribes (Enkoko) and still impaired in the binocular field as indicated H. abyssinica (Kosso), cause defects in visually by their performance on the visual discrimiguided behavior in chicks. This result is clearly nation task. Treatment with lower doses apevident in the reduced ability of treated chicks pears to spare the binocular field. We therefore to detect a small bead moving into the pe- conclude that these anthelmintic agents can ripheral field of vision. The visual field con- affect both the frontal and peripheral fields of stricts to no greater than 20 to 35” for chicks vision, but that the area of the retina processing treated with a total of 1.25 g (0.25 g X 5 days) information from the peripheral field is more and 0.225 g (0.025 g X 9 days) of Enkoko, and susceptible to the drugs. 2.25 g (0.25 g X 9 days) of Kosso. It is interA maximal reduction in ability to detect a esting to note that this size of the visual field small moving object, as well as impaired visual corresponds to the calculated binocular field discrimination performance, was found in of chicks, which is between 30 to 33” (Petti- chicks treated with a total amount of 0.009 g grew, 1983). Treatment with lower doses of of embelin. This suggests that embelin, an the drugs constricted the visual field to a lesser ether extract of E. ribes (Fieser and Chamextent. Thus, dose-dependency was evident berlin, 1948; Gupta et al., 1976, 1977; Chauboth in the rate of constriction of the visual han et al., 1979; Abdul Rasheed et al., 1983), field over days of treatment and in the final may be the active principle responsible for the size of detection angle reached. deficits in visually guided behavior in the EnThe significant narrowing of visual fields in koko-treated birds. The occurrence of embelin treated animals suggests that peripheral vision in E. ribes has been reported to be no more is affected. Two explanations can be put for- than 3% (Fieser and Chamberlin, 1948; Chauward to account for this deficit in vision. First, han et al., 1979; Abdul Rasheed et al., 1983). it is possible that the peripheral retina is af- On the basis of a yield of 3%, one can calculate fected, but the area that subserves binocular than a total of 0.009 g of embelin corresponds vision is spared. Alternatively, all areas of the to 0.27 g of dry weight of Enkoko. Incidentally, retina may be affected to a similar extent, but this amount of Enkoko (0.25 g) produced sigthe probability of detection is increased in the nificantly impaired performance on both tasks binocular field since both eyes are better than (see Figs. 3a and 5). Although a comparison one in the detection of the bead advancing between the effects of the embelin (0.001 g from the periphery to the frontal field of vision. X 9 days) and Enkoko (0.25 g X 1 day) groups Visual discrimination was also impaired af- may not be appropriate since the time course ter treatment with total amounts of 0.200 g of administration was different, it is interesting

228

LOW

ET

AL.

to note that both of these groups which re- visual discrimination, and optomotor response ceived effectively a similar amount of embelin tasks (unpublished results). exhibited correspondence in the degree of viWe cannot yet answer whether the retinosual deficits. toxicity caused by Enkoko or Kosso is caused Morphological changes in ganglion cells by the same active constituent(s) which confer were evident in the Enkoko (0.25 g X 5 days)- their anthelmintic and/or contraceptive propand Kosso (0.25 g X 9 days)-exposed retinas. erties. Enkoko has two constituents, embelin Doses of 0.25 g and lower of either Enkoko and vilangin (Thomson, 1971). Embelin from or Kosso, or 0.009 g of embelin, however, E. ribes has been extensively used and studied failed to cause pyknosis of ganglion cells. Nev- because of its anthelmintic (Harris, 1887; Parertheless, behavioral deficits were still present. anjpe and Gokhale, 1932; Mukerji and BhadIt is conceivable that the behavioral tests were uri, 1947; Guru and Mishra, 1964; Pate1 et al., able to detect visual deficits at doses lower than 1964; Pankhurst and Pearson, 1972: Dube et those for which morphological changes could al., 1982) and antifertility properties (e.g. be observed, a finding which is consistent with Radhakrishnan and Alam, 1975; Seshadri et other retinotoxicity studies (Rogers et al. al., 1978, 1979; Pankhurst, 1979; Prakash and 1984). It is interesting to note that patients Mathur, 1980; Dixit and Bhargava, 1983). The with Enkoko or Kosso poisoning had de- pharmacolgical action(s) of vilangin is as yet creased visual fields a few days after ingestion unknown. In the case of Kosso, it has long of these anthelmintics, but optic atrophy was been used as an anthelmintic drug (Pankhurst not observed until 4 to 6 weeks later (Rokos, and Pearson, 1972; Pankhurst, 1975, 1979) 1969). Therefore, retinopathy may have be- and is attributed to possess contraceptive-like come evident in the chicks treated with lower activity (Casey, 1960; Sareen et al., 196 1; doses if they had been older when their retinas Chaudhury, 1966; Farnsworth et al., 1975a,b). were sampled. The absence of clear morphoKosso is known to consist of three (Birch and logical changes in embelin-exposed retinas is Todd, 1952), or possibly four (Lounasmaa et not unexpected since the total dose (0.009 g: al., 1973) principles. One of these principles, 0.001 g X 9 days) of embelin is equivalent to kosotoxin, is anthelmintically active (Lounthe low dose (0.25 g: 0.25 g X 1 day) of En- asmaa et al., 1973). koko; retinas from chicks treated with this The constituents of Kosso and that of Endose of Enkoko did not show any retinotoxic koko are phloroglucinol derivatives (Lounaseffects. In vitro studies of eyeballs incubated maa, et al., 1973, 1974a,b; Joshi and Kamat, with embelin have revealed a severe loss of 1975; Lounasmaa and Varenne, 1978) with ganglion cells and the optic fiber layer (Low structural similarities to the toxin found in et al., 1985). Embelin could explain the ret- male fern, Dryopteris @ix-mas (Polypodiinotoxicity of Enkoko, but not KOSSO,since it aceae). This toxin is well known to produce is not a constituent of the latter. It is also im- retinal degeneration in animals (Macleod and portant to note that embelin is probably not Greig, 1978) and humans (Wirth and Gloxthe sole component of Enkoko responsible for uber, 1967; Grant, 1974, 1980). It is interesting the visual deficits. Rao and Venkateswarlu to note that the retinal pathology caused by ( 196 1) have isolated, and subsequently syn- male fern is restricted to death of ganglion cells thesized (Rao and Venkateswarlu, 196 1, and degeneration of optic nerve fibers (Grant, 1962) a minor constituent from E. ribes. 1974, 1980). It remains to be seen whether the which they called vilangin. Preliminary ex- constituents which cause the observed defects periments with dioxan extraction, which pre- in visual behavior and subsequent retinopasumably yields vilangin (Rao and Venkatesthology have anthelmintic and contraceptivewarlu, 196 1), causes significantly impaired like properties as well, or whether these desired performance in chicks on the visual detection, actions can be separated from the retinotox-

RETINOTOXICITY

CAUSED

icity. Most importantly, if indeed the retinotoxic components of these plants are found to have antifertility activity, their use as potential contraceptive agents must therefore be treated with some caution. ACKNOWLEDGMENTS We thank the Coordinating Office for Traditional Medicine (Ethiopia) for the supply of Enkoko and Kosso, and Dr. S. P. Poph (Central Drug Research Institute, Lucknow, India) for the generous supply of crystalline embelin. We also thank Mrs. S. Simpson for the artwork, and Mr. T. Martin for the photography. This study was supported by a grant from the Rengland Anderson Opthalmic Fund.

REFERENCES ABDUL RASHEED, K. K., CHACKO, J., AND NAMBISAN, P. N. K. (1983). Thermal, spectral and magnetic studies on some transition metal complexes of embelin. Polyhedron

2,293-299.

BIRCH, A. J., AND TODD, A. R. (1952). Anthelmintics: Kosso. Part II. The structures ofprotokosin, -kosin, and -kosin. J. Chern. Sot., 3102. CASEY, R. C. D. (1960). Alleged anti-fertility plants of India. Indian J. Med. Sci. 14, 590-600. CHAUDHURY, R. R. (1966). Plants with possible antifertility activity. Indian Council of Medical Research Special Report Series No. 55. CHAUHAN, S., AGRAWAL, S., MATHUR, R., ANDGUPTA, R. K. (1979). Phosphatase activity in testis and prostate of rats treated with embelin and vinca rosea extract. Experientia 35, 1183-l 185. DIXIT, V. P., AND BHARGAVA, S. K. (1983). Reversible contraception like activity of embelin in male dogs. (Canis indicus Linn). Andrologia 15,486-494. DUBE, C. B., BARNWAL, A. K., BAJPAI, D. K., AND KANSAL, C. M. (1982). Ascaris and vidanga (An Experimental Study). J. Sci. Res. Plants Med. 3,44-48. EHRLICH, D. (198 1). Regional specialization of the chick retina as revealed by the size and density of neurons in the ganglion cell layer. J. Comp. Neural. 195,643-657. FARNSWORTH, N. R., BINGEL, A. S., CORDELL, G. A., CRANE, F. A., AND FONG, H. H. S. (1975a). Potential value of plants as sources of new, antifertility agents I. J. Pharm.

Sci. 64,535-598.

FARNSWORTH, N. R., BINGEL, A. S., CORDELL, G. A., CRANE, F. A., AND FONG, H. H. S. (1975b). Potential value of plants as sources of new antifertility agents, II. J. Pharm.

Sci. 64,7

17-754.

FIESER,L. F., AND CHAMBERLIN, E. M. (1948). Synthesis of embehn, rapanone and related quinones by peroxide alkylation. J. Amer. Chem. Sot. IO, 7 l-75.

229

BY ANTHELMINTICS

GRANT, W. M. (1974). Toxicology of the Eye, 2nd ed. Thomas, Springfield, Illinois. GRANT, W. M. (1980). The peripheral visual system as a target. In Experimental and Clinical Neurotoxicology (P. S. Spencer and H. H. Schaumberg, eds.). Williams and Wilkins, Baltimore London. GUFTA, 0. P., ANAND, K. K., MOHD ALI, M., RAY GHATAK, B. J., AND ATAL, C. K. ( 1976). In vitro anthelmintic activity of disalts of embelin. Indian J. Expt. Biol. 14, 356-357. GUPTA, 0. P., MOHD. ALI, M., RAY GHATAK, B. J., AND ATAL, C. K. (1977). Some pharmacological investigations of embelin and its semisynthetic derivatives. Indian J. Physiol.

Pharmacol.

21, 31-39.

GURU, L. V., AND MISHRA, D. N. (1964). Anthelmintic activity of Embelia ribes (Burm) (Vayavirang) in vitro. Antiseptic

61, 545-550.

HARRIS, G. H. (1887). Lancet 2, 199. JOSHI, B. S., AND KAMAT, V. N. (1975). Benzoquinone derivatives. Part I. Reactions of primary aliphatic amines with embelin (2,5dihydroxy-3-undecyl-1,4benzoquinone) and Di-o-methylembeline. J. Chem. Sot. Perkin Tram

l., 327-332.

KEPPEL, G. (1982). Design and Analysis: A Researcher’s Handbook. Prentice Hall, Englewood Cliffs, New Jersey. LOUNASMAA, M., WIDEN, C. J., AND HUHTIKANGAS, A. (1973). Phloroghrcinol derivatives of Hagenia abyssinica. Phytochemistry

12, 2017-2025.

LQUNASMAA, M., WIDEN, C. J., AND HUHTIKANGAS, A. (1974a). Phloroghtcinol derivative of Hagenia abyssinica. Part 2. The structure determination of koso toxin and proto kosin. Acta. Chem. Stand. Ser. B. Org. Chem. Biochem. 28, 1200- 1208. LOUNASMAA, M., WIDEN, C. J., AND HUHTIKANGAS, A. (1974b). Phloroglucinol derivative of Hagenia abyssinica. Part 3. Reductive alkaline cleavages of koso toxin and proto kosin and of Aspidin BB Dryopteris-Assimihs. Acta.

Chem.

Stand.

Ser. B. Org. Chem.

Biochem.

28,

1209-1218. LOUNASMAA, M. AND VARENNE, P. (1978). Derives Phloroglucinoligues d’Hagenia alyssinica V. Planta Medica 34, 153-159. Low, G., ROGERS,L. J., BRUMLEY, S., AND EHRLICH, D. (1985). Retinotoxic effects of Embelia species(Enkoko) on avian and mammalian visual systems.Neurosci. Lett. (Suppl.) 19, S80. MACLEOD, N. S., AND GREIG, A. (1978). Poisoning in cattle associated with Dryopterisfrlix-mas & D. borreri. Vet. Rec. 102, 239-240.

MERIGAN, W. H. (1980). Visual fields and flicker thresholds in methylmercury-poisoned monkeys. In Neurotoxicity of the Visual System (W. H. Merigan and B. Weiss, eds.), pp. 149, 150. Raven Press, New York. MUKEFUI, A. K., AND BHADURI, N. V. (1947). The treatment of intestinal worms with the indigenous drugs, Butea, Embelia and Kamala. Ind. Med. Gaz. 82, 66-69.

230

LOW ET AL.

PANKHURST,R. (1975). Historical Anecdote: Dr. A. Brayer & Europe’s Discovery of Kosso. Ethiop. Med. J 13,2934. PANKHURST, R. ( 1979). Europe’s discovery of the Ethiopian taenicide - Kosso. Medial History 23,297-3 13. PANKHURST, R., AND PEARSON, T. (1972). Remedius Prutky’s 18th Century account of Ethiopian taenicides and other medical treatment. Ethiop. Med. J. 10, 3-6. PARANJPE, A. S., AND GOKHALE, G. K. (1932). Pharmacological study of embelin with reference to its use as an anthelmintic. Arch. Int. Pharmacodyn. 42, 212232. PATEL, R. P., SHAH, C. S., KHANNA, P. N., ANDGHANDI, T. P. (1964). Pharmacognostic and pharmacological studies of Embelia ribes Burm and Embrebia tsyeriamCotton A-DC (Syn. E. Robusfa Parke).Indian J. Pharm. 26, 168-172. PETTIGREW,J. D. (1983). The sensory space of birds. In Proceedings

18th International

Ethological

Conference,

p. 228. POTTS, A. M. (I 980). Duality of the optic nerve in toxicology. In Neurotoxicity of the Visual System (W. H. Mergian and B. Weiss, eds), pp. 1,2. Raven Press, New York. PRAKASH, A. O., AND MATHUR, R. (1980). Effect of few indigenous plant extracts on corpora lutea of cyclic guinea pigs. Bioresearch (Uljain) 4, 35-40. PURANDARE,T. V., KHOLKUTE, S. D., GURJAR,A., JOSHI, U. M., DA~ATREYAMURTY, B., SHETH,A. R., SWAMY, X. R., JAYARAMAN, S., AND MUNSHI, S. R. (1979). Semen analysis and hormonal levels in Bonnet Macaques administered Embelia ribes berries, an indigenous plant having contraceptive activity. Indian J. Exp. Biol. 17, 935-936. RADHAKRISHNAN,N., AND ALAM, M. ( 1975). Antifertility activity of embelin in albino rata. Indian J. Exp. Biol. 13,70-71. RAO, CH.B., AND VENKATESWARLU, V. (1961). Chemical examination of Embelia ribes. I. Isolation of a new constituent, ‘viIangin’. Its constitution and synthesis. J. Org. Chem.

26,4529-4532.

RAO, CH.B., AND VENKATESWARLU,V. (1962). Chemical examination of Embelia ribes. III. A new synthesis of vilangin. Tetrahedron 18,95 1. RODIECK, R. W. (1973). The Vertebrate Retina. Principles of Structure and Function. Freeman, San Francisco. ROGERS, L. J., DRENNEN, H., AND MARK, R. F. (1974). Inhibition of memory formation in the imprinting period: Irreversible action of Cxm in young chickens. Brain Res. 74, 2 13-233. ROGERS, L. J., EHRLICH, D., AND ZAPPIA, J. V. (1984). Visual defects in chickens with kainic acid and glutamate induced lesions of the retina. Proc. Aust. Physiol. Pharmacol. Sot. 15, 38. ROKOS, L. (1969). Eye complications in poisoning caused by ‘Koso’ (Hagenia abyssinica). Ethiop. Med. J. 7, 1 l16. SAREEN,K., MISRA, N., VARMA, D. R., AMMA, M. K. P., ANDGUJRAL, M. L. (1961). Oral contraceptives - Part V. Anthelmintics as antifertility agents. Indian J. Physiol. Pharmacol. 5, 125-l 35. SESHADRI,C., SUGANTHAN, D., SA NTHAKUMARI, G., AND IYER, G. Y. N. (1978). Biochemical changes in the uterus and uterine fluid of mated rats treated with embelin-a non-steroidal oral contraceptive. Indian J. Exp. Biol. 16, 1187-l 188. SESHADRI, C., SUGANTHAN, D., AND SANTHANKUMARI, G. (1979). Biochemical changes in the uterus and cervix of ovariectimized rats treated with embelin. Indian J. Exp. Biol. 17, 1389-1390. THOMSON, R. H. (197 1). Naturally Occurring Quinones, p. 137. Academic Press, New York. WALLMAN, J., AND TURKEL, J. (1978). Extreme myopia produced by modest change in early visual experience. Science (Washington, D.C.) 201, 1249- 125 1. WINER, B. J. (197 1). Statistical Principles in Experimental Design, 2nd ed. McGraw-Hill, New York. WIRTH, W.. AND GLOXUBER, Ch. (1967). Toxikologie, p. 283. Fibel. Georg. Threine Verlag, Stuttgart. YINON, U., ROSE, L., AND SHAPIRO, A. (1980). Myopia in the eye of developing chicks following monocular and binocular lid closure. Vision Res. 20, 137- 14 1.